& real Estate cycles - Adventures in CRE

[Pages:16]Cornerstone Real Estate Advisers LLC

cap rates & real Estate

cycles:

A historical perspective with a look to the future

Jim clayton, ph.D., Vice President, Research Cornerstone Real Estate Advisers LLC 860.509.2237

lisa Dorsey glass, Managing Director, Babson Capital Management LLC lrglass@

June 2009

Introduction

Capitalization or "cap" rates play a central role in real estate investment, financing and valuation decisions. Average market-wide cap rates are widely quoted and followed as a gauge of current real estate investment market conditions. Cap rates received increasing attention in both industry and academic circles over the past decade, as real estate established itself as a mainstream asset class that became more integrated with broader capital markets, on both the debt and equity sides. The resulting surge of capital into the real estate sector over the last decade helped drive property values to historical highs and cap rates to new lows. The phrase "cap rate compression" was born as cap rates fell from the 8-10% range in the early 2000's to 5-7% by 2006 (Exhibits 1 and 2). During this period, and especially the later part of it, the appropriate level of cap rates was widely discussed and debated amongst the "new paradigm-real estate risk has been permanently re-priced" and "pricing bubble" camps. Today, as the real estate sector works its way through a deep financial crisis-induced recession, cap rates are increasing and investors are struggling to get a handle on just how high they will go and where they will settle once a new equilibrium is reached. Moreover, in today's environment, characterized by limited transaction activity, market derived information about cap rates is not widely available.

ExhIbIt 1: InstItutIonAl propErty cAp rAtEs

(Quarterly, 1990:1 - 2009:1)

12% Apartment

11%

Retail

Industrial

Office

10%

9%

8%

7%

6%

5%

4% 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08

"Current Value" cap rates indicative of the trend in appraisal cap rates. Source: NCREIF

ExhIbIt 2: AvErAgE trAnsActIon cAp rAtEs - All InvEstor typEs

(Monthly, Jan. 2001 - Feb. 2009)

11% 10%

Apartment

Industrial

Office-CBD

Office-Suburb

Retail-Strip Ctr

9%

8%

7%

6%

5% 2001

2002

2003

2004

2005

Derived from property transactions of $5 million and greater.

Source: Real Capital Analytics

2006

2007

2008

2009

One Financial Plaza, Suite 1700 Hartford, CT 06902 1

cornerstone Research

While cap rates are widely quoted and referenced, there remains considerable confusion about and misinformation related to exactly how they are determined and what they mean. This paper aims to fill this knowledge gap and provide readers with a sound understanding of both the conceptual underpinnings and the fundamental determinants of cap rates. The paper also examines cap rate dynamics during previous recessionary periods with the intent of gaining a better understanding of cap rate behavior during the current economic downturn and what we might expect looking forward as the current economic cycle plays out.

WhAt Is A cAp rAtE? At a fundamental level, overall capitalization, or "cap", rates are a way of quoting observed property prices in relation to expected first year asset-level income. A cap rate is essentially the expected first year income yield on an income property investment. It is defined as the ratio of property net operating income (NOI) to current market value (V), or

cap rate = NOI1

(1)

V

In the realm of commercial real estate, the capitalization rate is a tool widely used to estimate the value of a particular property. It is the foundation of the "direct capitalization" method of real estate valuation. In this context, income-property can be valued by applying a cap rate to an estimate of first year net operating (NOI). That is, the above expression for the cap rate can be arranged to yield,

V = NOI1 cap rate

(2)

For example, if the appropriate cap rate for an office building producing an annual NOI of $1 million is 10%, then the estimated value of the property is $10 million. A lower (higher) cap rate would imply a higher (lower) property value; there is an inverse relationship between cap rates and value assuming a static income stream. This valuation approach assumes, of course, we know the cap rate.

WhErE Do cAp rAtEs comE From? Cap rates are generally derived from observed property transactions.1 Most institutional investors estimate property values with a discounted cash flow (DCF) methodology, using a multi-year pro forma and taking the present value (PV) of expected future cash flows (CFs), including expected net sale or reversion proceeds (REV) at the end of an assumed "T" year holding period, discounted at the appropriate risk-adjusted required total return k.2 That is, property value (V) is determined as

V

=

CF1 (1+ k)

+ CF2 (1 + k)2

+ CF3 (1 + k)3

+ ...+ CFT + REVT (1 + k)T

(3)

1. In theory cap rates can also be constructed as the weighted average of typical investor's required first year equity returns and the cost of debt, assuming a typical or average loan to value ratio. This approach is known as the band-of-investment method of building up a cap rate.

2. k is also termed the discount rate or the opportunity cost of capital (OCC) or the unlevered IRR.

Cornerstone Real Estate Advisers LLC 2

cornerstone Research

and the "going-in cap rate" or first year income return on assets, is defined as CF1/V. Note that we have switched from net operating income (NOI) to the more general cash flow (CF) that may include an annual reserve for future capital improvements, leasing costs and tenant improvement expenditures. The cap rate provides a summary measure of price paid per dollar of expected first year property income and implicitly includes the impact of leasing and tenant improvement expenditures. In an active market, cap rates extracted from recently completed transactions can provide investors and appraisers a useful guide for determining the appropriateness of the cap rate to be used in valuing a subject property.

cAp rAtE DAtA AnD thE cyclIcAl bEhAvIor oF cAp rAtEs The cap rate series in Exhibits 1 and 2 come from two widely referenced sources. Exhibit 1 shows average cap rates derived from appraisals of core institutional properties included in the benchmark property return index ("NPI") produced quarterly by the National Council of Real Estate Investment Fiduciaries ("NCREIF"), dating back to 1990. The NPI consists of quarterly performance data for unlevered investment-grade properties owned by or on behalf of taxexempt institutions such as pension funds, endowments and foundations. Income producing assets from the major property types are included in the index: apartments, industrial, office, and retail properties; hotels are excluded. Exhibit 2 displays monthly series of average transaction cap rates reported by Real Capital Analytics ("RCA") dating back to 2001. RCA data is derived from a broader sample of properties compared to NCREIF as the data attempts to cover all transactions of $5 million or more, of which institutional transactions are one component.

RCA cap rate data does not exist prior to 2001. The NCREIF Property Index began in 1978 and therefore allows us to examine the behavior of cap rates, albeit only on the larger core properties owned by institutional investors, over the past three decades. In what follows we study aggregate or average NCREIF cap rates, as opposed to the property type level, since this allows us to go further back in time. In addition, we examine what NCREIF terms "current value" cap rates, reflecting cap rates for recently appraised properties, and also "transaction" cap rates derived from the sales of properties included in the NCREIF index. Ideally, we would want to focus on cap rates derived from transactions to provide the most up to date information about pricing. However, the NCREIF transaction cap rate series does not begin until 1983 and in the early years is quite erratic as the figures are derived from a relatively small number of transactions. In addition there is considerable divergence in current value and transaction cap rates through much of the 1980s and into the early 1990s.

Exhibit 3 displays the NCREIF cap rate time series, with appraisal cap rates dating back to 1979 and transaction cap rates dating back to the early 1990s. It clearly highlights the cyclical nature of real estate investment markets. Cap rates vary over time as macroeconomic conditions and real estate space and capital markets fluctuate. Property income and expectations of future growth vary with economic and local supply / demand fundamentals. The rate of capitalization of

Cornerstone Real Estate Advisers LLC 3

cornerstone Research

property income into property value depends also on capital market conditions as reflected in the opportunity cost of capital and risk perception associated with the real estate asset class. For much of the 1980's, cap rates declined and real estate prices trended upward as the move by pension funds, Japanese investors and other institutions into real estate coincided with aggressive lending and a subsequent period of significant overbuilding. This precipitated a sharp run up in commercial property values. Inflation fears and institutional investor demand bolstered the commercial real estate market at a time when changes in tax laws enhanced already generous depreciation allowances and tax shelters for wealthy individuals. Also during the 1980's, the deregulation of the savings and loan ("S&L") industry allowed these institutions to originate and/or invest in commercial mortgages for the first time, thus opening up another new and inexperienced source of capital for commercial property, adding fuel to the real estate boom. By the late 1980's, tax reform had eliminated most of the special tax incentives. The overbuilding and extent of the financial crisis in the S&L industry were realized; and subsequently the commercial property market suffered a major downturn characterized by significant declines in real estate prices resulting in much higher cap rates. Over the 2003?07 period, dramatic increases in debt and equity availability, improving property fundamentals, and increased investor demand produced material declines in cap rates and increases in prices. Most recently (2007-2009) unprecedented contractions of the credit markets, lower investor demand and softening property fundamentals have resulted in an ascent of cap rates that is expected to continue as the economy navigates through one of the most severe recessions on record. While our focus in this paper will be on average sector wide pricing, Exhibits 1 and 2 show cap rates do vary across property types due to variations in property income growth prospects and risk premiums.

ExhIbIt 3: ncrEIF ApprAIsAl AnD trAnsActIon cAp rAtEs

12% 11% 10%

Current Value (Appraisal) Cap Rate Transaction Cap Rate (Sold Properties) Average Current Value Cap Rate

Cap Rates

9%

8%

? 1 SD

7%

6%

Average 7.62%

5%

4% 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09

Source: NCREIF

Exhibit 3 reveals that while institutional property cap rates do vary over time, they do so in a somewhat predictable manner, never getting too far from their long-run average of 7.6 % (the red line in Exhibit 3). Moreover, with the exception of the period of the mid-nineties and the most recent property price upswing, current value cap rates remain within a band of 6.75% to 8.75% (the shaded area

Cornerstone Real Estate Advisers LLC 4

cornerstone Research

in Exhibit).3 The period since 2005 stands out as a somewhat unique one for cap rates, with both current value and transaction cap rates "falling" well out of the historical range.

lookIng InsIDE cAp rAtEs The basic property valuation formula in equation (3), reveals, in addition to first year expected property income, a cap rate is a function of the discount rate, k, and expected change in future cash flows. All else being equal, the higher the property income growth expectations, the lower the cap rate. Investors are willing to pay more per dollar of current income (CF1) the larger the expected cash flows in subsequent years. Under certain simplifying assumptions we can derive an exact relationship between the discount rate, the expected rate of growth of property cash flows and the cap rate.4 More generally, we can show the following approximate relationship holds and is very useful in terms of understanding the determinants of cap rates and the dynamics of changes in cap rates:

cap rate k - g

(4)

We can go further and break the required total return, k, into two components,

namely the risk-free rate, kRF, (or yield to maturity on default-free government bonds, generally proxied by the yield on 10 year Treasuries) plus a real estate

risk premium, RP, so that

cap rate (kRF + RP ) - g

Cap Rate Level Determinants (5)

This decomposition nicely illustrates the three key determinants of cap rates:

Yield on government bonds (kRF) Real estate risk premium (RP) Property income growth expectations (g)

Variation in cap rates over time is driven by changes in the opportunity cost of capital as reflected in the yield to maturity on default risk-free (government) bonds and the risk premium added to the bond yield, as well as fluctuations in the expected long-term growth of property cash flows. In thinking about the components of variation in cap rates over time, it is important to note that kRF comes from the capital markets, RP comes from both the capital markets and the real estate space markets, while g comes from the real estate space or user market. We would expect movements in bond yields and revisions in property

3. See "The Long Cycle in Real Estate." By Ron Kaiser in the Journal of Real Estate Research, vol. 14, no. 3 (1997) for evidence that cap rates have generally remained within this range dating back to the early 1950s.

4. Under certain assumptions, the present value property valuation formula in equation (3) collapses to a nice simple formula that is often used to highlight the key determinants of cap rates. In the special case where we expect to hold the property indefinitely (i.e. T goes to infinity), and annual operating cash flows are expected to grow at a constant rate "g", equation (1) becomes

V

=

CF1 (1+ k)

(

+

1

+

g)CF1

+(1

+

g

)

2CF1

(1+ k)2

(1 + k)3

+

=

CF1

(1

1 +

k)

+

(1+ g )

(1+ k)2

+(1+ g)2

(1+ k)3

+ ...

=

CF1

k

1 -

g

In this case, the relationship cap rate = k ? g holds exactly.

Cornerstone Real Estate Advisers LLC 5

cornerstone Research

income growth (rents and vacancy rates) would be the most volatile components of cap rate movements, with the risk premium expected to change more slowly over time, especially at the aggregate or sector wide level. It is also crucial to recognize the three cap rate ingredients are not independent of one another; cap rate fluctuations result from a complex interaction of the three variables. One cannot say, for example, cap rates move one to one with Treasury yields ? it depends on what drove the change in Treasury yields and what this implies for growth expectations and the risk premium.

Investors often quote real estate cap rates or yields in terms of spreads above Treasuries, and sometime refer to this spread as a risk premium. Subtracting the risk-free rate from both sides of Equation (5) yields the following expression for the cap rate ? Treasury spread:

(cap rate - kRF ) (RP - g )

Cap rate spread to 10 yr. Tsy yield

Cap Rate Spread Determinants (6)

The spread is positively related to the risk premium, but also depends crucially on property income growth expectations; it is too simplistic to call the spread between cap rates and Treasury yields a risk premium. In theory, it is possible to have a negative spread during periods of high property income growth expectations. Exhibit 4 reveals this situation did in fact persist for much of the 1980s, a period preceding the major real estate downturn in the early 1990s. Subsequent to the recovery of the early 1990's downturn, the spread between institutional cap rates and 10 year Treasury yields has remained positive. Cap rates moved and remained high in the early 1990's as Treasuries trended downwards, and then eventually began to follow Treasury yields down. The cap rate-Treasury spread remained relatively wide until 2003 when cap rates continued to compress at a rapid clip even as 10 year government bond yields bottomed out resulting in a sharp contraction of the spread, especially in 2006 and 2007 as Treasury yields began to move up.

ExhIbIt 4: ncrEIF trAnsActIon cAp rAtEs AnD 10 yEAr trEAsury yIElDs

14% 12%

Transaction Cap Rate

10 Yr. Treasury Yield

10%

8%

6%

4%

2%

0% 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 1Q09

Source: Federal Reserve, NCREIF

Cornerstone Real Estate Advisers LLC 6

cornerstone Research

As the decline in cap rates that began in 2002 continued, there was considerable debate about whether this was due to a re-pricing of the real estate asset class which was now being acknowledged as a more mainstream accepted asset class, or a short-term capital flow phenomenon with too many investors chasing too little product and bidding prices up. Research undertaken early into the 20022007 real estate upswing generally concluded that rising property valuations could be explained by market fundamentals and rational pricing of real estate relative to other asset classes.5 The re-pricing of risk and resulting reductions in risk premiums and spreads was an economy wide phenomenon affecting all risky asset classes not just real estate. Exhibit 5 shows yields on Baa-rated corporate bonds trended downwards with 10 year Treasury yields over most of the post 1990 period, and the Baa-10 yr. Treasury spread also declined significantly starting in 2003, before spiking in the credit crisis. Baa corporate bonds generally traded inside of (i.e. at lower yields than) cap rates since 1992, due in part to their higher level of liquidity.

ExhIbIt 5: cAp rAtEs AnD bonD yIElDs

%

12

10 yr Treasury

Corporate Baa

11

10

9

8

7

6

5

4

3

2 1990

1992

1994

1996

1998

2000

Source: Federal Reserve, NCREIF and Cornerstone Research

2002

NCREIF Cap Rate

2004

2006

2008

The relationship between real estate cap rates and corporate bond yields inverted in 2005; which possibly suggests real estate pricing became too aggressive. Corporate bond yields spiked as the grip of the credit crisis took hold, while adjustment in cap rates initially lagged behind. It appears that a more normal situation is on the horizon as cap rates increase and bond yields move down, albeit slowly. Easy access to low cost debt may have helped fuel a surge in real estate equity investment activity that took property investment markets into unsustainable territory. Abundant liquidity in the debt markets indirectly fueled pricing competition in investment sales with investors underestimating risk and/or overestimating growth in property income and appreciation.6 As part of

5. See for example, "Real Estate Pricing: Spreads and Sensibilities: Why Real Estate Pricing is Rational," J. Chen, S. Hudson-Wilson and H. Nordby, Journal of Real Estate Portfolio Management, 2004.

6. Two recent studies conclude that "investor sentiment" helped push property values too high (cap rates too low) relative to underlying intrinsic value. See "Commercial Real Estate Valuation: Fundamentals versus Investor Sentiment," by J. Clayton, D. Ling and A. Naranjo, Journal of Real Estate Finance and Economics, 2009 and "The Secular and Cyclic Determinants of Capitalization Rates: The Role of Property Fundamentals, Macroeconomic Factors, and `Structural Changes'" by S. Chervachidze, J. Costello and W. Wheaton, forthcoming in the Journal of Portfolio Management special Real Estate issue, 2009.

Cornerstone Real Estate Advisers LLC 7

cornerstone Research

the so-called global "wall of capital," aggregate commercial and multi-family mortgage debt surged over the past decade. From just over $1 trillion in 1997 as the sector finally emerged from the early 1990's downturn, to $3.4 trillion in the third quarter of 2008, taking the size of the mortgage market from just shy of 14% of GDP to nearly 25% (Exhibit 6), topping the 1980's peak that was just less than 20% of GDP.7

Much of the recent increase in leverage in the commercial real estate sector was attributable to unbelievable (we now know "unsustainable") growth in commercial mortgage backed security ("CMBS") issuance that in 2006 and 2007 was characterized by increasing complexity in security structure and funding sources, weaker underwriting, and/or overconfidence on the part of both lenders and bond rating agencies. The result was a proliferation of ever riskier highleverage loans on frothy property valuations backing CMBS bonds. The fallout has helped push the commercial property sector into a painful period of de-leveraging and balance sheet repair-induced illiquidity and value decline.

Mortgage Debt as a % of GDP NCREIF Current Value Cap Rate

ExhIbIt 6: cAp rAtEs AnD mortgAgE DEbt outstAnDIng

24% 22%

Comm & Multi Fam Mtg Debt as % of GDP

NCREIF Cap Rate

10.0% 9.5% 9.0%

20% 18% 16%

8.5% 8.0% 7.5% 7.0%

14% 12%

6.5% 6.0% 5.5%

10%

5.0%

1978Q4 1980Q4 1982Q4 1984Q4 1986Q4 1988Q4 1990Q4 1992Q4 1994Q4 1996Q4 1998Q4 2000Q4 2002Q4 2004Q4 2006Q4 2008Q4

Source: Cornerstone Research based on NCREIF, Bureau of Economic Analysis and Federal Reserve data

cAp rAtE DynAmIcs In prEvIous rEcEssIons The previous sections provided a conceptual framework for understanding the determinants of cap rates and the major factors causing them to change over time. To gain an even better understanding of cap rate behavior, this section provides a more detailed historical account of cap rate trends in periods surrounding past recessions, with the aim of uncovering any systematic cap rate behaviors associated with the movement of the broader economy into and out of recessions. This analysis will be limited to recessions as defined by the National Bureau of Economic Research ("NBER") which is considered the official authority for dating U.S. recessions. Rather than defining a recession by the school book measure of two or more consecutive quarters of economic contraction as measured by

7. Both the Clayton et al. and Chervachidze et al. studies referenced in the previous footnote relate investor sentiment to debt as a percent of GDP.

Cornerstone Real Estate Advisers LLC 8

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download